Purpose: This study explored whether measures of working memory ability contribute to the wide variation in 2-year-olds' expressive vocabulary skills. Method: Seventy-nine children (aged 24-30 months) were assessed by using standardized tests of vocabulary and visual cognition, a processing speed measure, and behavioral measures of verbal working memory and phonological short-term memory. Results: Strong correlations were observed between phonological short-term memory, verbal working memory, and expressive vocabulary. Speed of spoken word recognition showed a moderate significant correlation with expressive vocabulary. In a multivariate regression model for expressive vocabulary, the most powerful predictor was a measure of phonological short-term memory (accounting for 66% unique variance), followed by verbal working memory (6%), sex (2%), and age (1%). Processing speed did not add significant unique variance. Conclusions: These findings confirm previous research positing a strong role for phonological short-term memory in early expressive vocabulary acquisition. They also extend previous research in two ways. First, a unique association between verbal working memory and expressive vocabulary in 2-year-olds was observed. Second, processing speed was not a unique predictor of variance in expressive vocabulary when included alongside measures of working memory.